As the ubiquity and complexity of software increase, a requirement has emerged for critical software that can successfully evolve without loss of quality – software that is engineered from the start to be easily changed, extended, and reconfigured, while retaining its security, its performance, and its reliability and predictability. The articles in this EssentialSet bring together the perspectives of key software engineering researchers and practitioners who both influence their organizations and evaluate the emerging practice of developing these new systems.
@proceedings{Hinchey2011ECS,
title = {Evolving Critical Systems: Essential Articles on Software Engineering},
year = {2011},
editor = {Hinchey, Mike and Coyle, Lorcan and Nuseibeh, Bashar and Fiadeiro, Jos{\'e} Luiz},
series = {EssentialSets: Anthologies from the Computer Society Digital Library},
publisher = {IEEE},
month = jun,
owner = {lorcan},
pdf = {pdf/2011/Hinchey2011ECS.pdf},
timestamp = {2011.06.27},
url = {http://www.computer.org/portal/web/store?product_id=ES0000036&category_id=TechSets}
}
Increasingly software can be considered to be critical, due to the business or other functionality which it supports. Upgrades or changes to such software are expensive and risky, primarily because the software has not been designed and built for ease of change. Expertise, tools and methodologies which support the design and implementation of software systems that evolve without risk (of failure or loss of quality) are essential. We address a research agenda for building software that (a) is highly reliable and (b) retains this reliability as it evolves, either over time or at run-time. We propose Evolving Critical Systems as an area for research to tackle the challenge and outline a number of scenarios to highlight some of the important research questions that should be asked of the community. Given that software evolution can be seen as a compromise between cost and risk, the most pressing question to ask is which processes, techniques and tools are most cost-effective for evolving critical systems?
@inproceedings{Hinchey2010ECSKeynote,
author = {Hinchey, Mike and Coyle, Lorcan},
title = {Evolving Critical Systems},
booktitle = {17th IEEE International Conference on the Engineering of Computer-Based
Systems},
year = {2010},
pages = {4},
address = {Los Alamitos, CA, USA},
month = mar,
publisher = {IEEE Computer Society},
note = {Keynote Presentation},
doi = {http://doi.ieeecomputersociety.org/10.1109/ECBS.2010.62},
isbn = {978-0-7695-4005-4}
}
Increasingly software can be considered to be critical, due to the business or other functionality which it supports. Upgrades or changes to such software are expensive and risky, primarily because the software has not been designed and built for ease of change. Expertise, tools and methodologies which support the design and implementation of software systems that evolve without risk (of failure or loss of quality) are essential. We address a research agenda for building software in computer-based systems that (a) is highly reliable and (b) retains this reliability as it evolves, either over time or at run-time.
@inproceedings{Hinchey2010ECSAgenda,
author = {Hinchey, Mike and Coyle, Lorcan},
title = {Evolving Critical Systems: A Research Agenda for Computer-Based Systems},
booktitle = {17th IEEE International Conference on the Engineering of Computer-Based
Systems},
year = {2010},
pages = {430-435},
address = {Los Alamitos, CA, USA},
month = mar,
publisher = {IEEE Computer Society},
doi = {http://doi.ieeecomputersociety.org/10.1109/ECBS.2010.56},
isbn = {978-0-7695-4005-4}
}
@article{Coyle2010ECS,
author = {Coyle, Lorcan and Hinchey, Mike and Nuseibeh, Bashar and Fiadeiro, Jos{\'e} Luiz},
title = {Guest Editors' Introduction: Evolving Critical Systems},
journal = {IEEE Computer},
year = {2010},
volume = {43},
pages = {28-33},
number = {5},
month = may,
doi = {10.1109/MC.2010.139},
owner = {lorcan},
timestamp = {2010.11.23}
}
@inproceedings{Knox2010Ontologies,
author = {Knox, Stephen and Coyle, Lorcan and Dobson, Simon},
title = {Using Ontologies in Case-Based Activity Recognition},
booktitle = {23rd Florida Artificial Intelligence Research Society Conference
(FLAIRS-23)},
year = {2010},
editor = {Guesgen, Hans W. and Murray, R. Charles},
month = may,
publisher = {AAAI Press},
owner = {lorcan},
pdf = {pdf/2010/Knox2010Ontologies.pdf},
timestamp = {2010.01.26}
}
Though computer scientists agree that conference publications enjoy greater status in computer science than in other disciplines, there is little quantitative evidence to support this view. The importance of journal publication in academic promotion makes it a highly personal issue, since focusing exclusively on journal papers misses many significant papers published by CS conferences. Here, we aim to quantify the relative importance of CS journal and conference papers, showing that CS papers in leading conferences match the impact of papers in mid-ranking journals and surpass the impact of papers in journals in the bottom half of the Thompson Reuters rankings (http://www.isiknowledge.com) for impact measured in terms of citations in Google Scholar. We also show that poor correlation between this measure and conference acceptance rates indicates conference publication is an inefficient market where venues equally challenging in terms of rejection rates offer quite different returns in terms of citations. How to measure the quality of academic research and performance of particular researchers has always involved debate. Many CS researchers feel that performance assessment is an exercise in futility, in part because academic research cannot be boiled down to a set of simple performance metrics, and any attempt to introduce them would expose the entire research enterprise to manipulation and gaming. On the other hand, many researchers want some reasonable way to evaluate academic performance, arguing that even an imperfect system sheds light on research quality, helping funding agencies and tenure committees make more informed decisions. One long-standing way of evaluating academic performance is through publication output. Best practice for academics is to write key research contributions as scholarly articles for submission to relevant journals and conferences; the peer-review model has stood the test of time in determining the quality of accepted articles. However, today’s culture of academic publication accommodates a range of publication opportunities yielding a continuum of quality, with a significant gap between the lower and upper reaches of the continuum; for example, journal papers are routinely viewed as superior to conference papers, which are generally considered superior to papers at workshops and local symposia. Several techniques are used for evaluating publications and publication outlets, mostly targeting journals. For example, Thompson Reuters (the Institute for Scientific Information) and other such organizations record and assess the number of citations accumulated by leading journals (and some high-ranking conferences) in the ISI Web of Knowledge (http://www.isiknowledge.com) to compute the impact factor of a journal as a measure of its ability to attract citations. Less-reliable indicators of publication quality are also available for judging conference quality; for example, a conference’s rejection rate is often cited as a quality indicator on the grounds that a high rejection rate means a more selective review process able to generate higher-quality papers. However, as the devil is in the details, the details in this case vary among academic disciplines and subdisciplines. Here, we examine the issue of publication quality from a CS/engineering perspective, describing how related publication practices differ from those of other disciplines, in that CS/engineering research is mainly published in conferences rather than in journals. This culture presents an important challenge when evaluating CS research because traditional impact metrics are better suited to evaluating journal rather than conference publications. In order to legitimize the role of conference papers to the wider scientific community, we offer an impact measure based on an analysis of Google Scholar citation data suited to CS conferences. We validate this new measure with a large-scale experiment covering 8,764 conference and journal papers to demonstrate a strong correlation between traditional journal impact and our new citation score. The results highlight how leading conferences compare favorably to mid-ranking journals, surpassing the impact of journals in the bottom half of the traditional ISI Web of Knowledge ranking. We also discuss a number of interesting anomalies in the CS conference circuit, highlighting how conferences with similar rejection rates (the traditional way of evaluating conferences) can attract quite different citation counts. We also note interesting geographical distinctions in this regard, particularly with respect to European and U.S. conferences.
@article{Freyne2010ConferenceVsJournals,
author = {Freyne, Jill and Coyle, Lorcan and Smyth, Barry and Cunningham, P{\'a}draig},
title = {Relative Status of Journal and Conference Publications in Computer
Science},
journal = {Communications of the ACM},
year = {2010},
volume = {53},
pages = {124-132},
number = {11},
month = nov,
doi = {10.1145/1839676.1839701},
owner = {lorcan},
pdf = {/pdf/2010/Freyne2010ConferenceVsJournals.pdf},
timestamp = {2010.11.23}
}
@article{McKeever2010Activity,
author = {Mckeever, Susan and Ye, Juan and Coyle, Lorcan and Bleakley, Chris and Dobson, Simon},
title = {Activity recognition using temporal evidence theory},
journal = {J. Ambient Intell. Smart Environ.},
year = {2010},
volume = {2},
pages = {253--269},
number = {3},
address = {Amsterdam, The Netherlands, The Netherlands},
issn = {1876-1364},
publisher = {IOS Press}
}
The rise of the Internet, the ever increasing ubiquity of data, and its low signal-to-noise ratio have contributed to the problem of information overload, whereby individuals have access to more data than they can assimilate into meaningful and actionable information. Much of the success of Web 2.0 has been achieved after an effective tackling of this problem. Ambient Information Systems take the battle into the physical world by integrating information into the physical environment tin a non-intimidating and non-overloading fashion. After two international workshops on Ambient Information Systems, we outline our vision for the field, consolidate a new definition, identify the key concerns of the research community, and issue a call to arms for future research.
@article{Hazlewood2009AIS,
author = {Hazlewood, William R. and Coyle, Lorcan},
title = {On Ambient Information Systems: Challenges of Design and Evaluation},
journal = {International Journal of Ambient Computing and Intelligence (IJACI)},
year = {2009},
volume = {1},
pages = {1-12},
number = {2},
owner = {lorcan},
pdf = {pdf/2009/Hazlewood2009AIS.pdf},
timestamp = {2011.08.12}
}
Detecting human activities from sensors deployed in the environment is a difficult problem. Human activities are so varied that it is difficult to match sensed activities to observed actions. Instrumented homes have proved a popular platform for such sensors but experiences there show how challenging it can be. We believe that the pool hall is a more useful platform for the deployment of body-mounted sensor networks as the motions that can be captured correlate well to the activities that take place on the pool table. The problem is more narrow than detecting human activities in general. We will focus on a simple sensor, mounted on the cue, which is important as the cue is the sole actuator used to play the game. By correlating the force with which a player strikes the ball with their cue and the distance the ball travels we demonstrate that with a simple off-the-shelf sensor we can derive useful information. We conclude by describing a set of enhancements that enable a computer to “sense” the game of pool.
@misc{Dalton2009Pool,
author = {Dalton, Deirdre and Coyle, Lorcan},
title = {Learn to Play like Minnesota Fats: Augmented Reality in the Pool
Hall},
month = sep,
year = {2009},
owner = {lorcan},
pdf = {pdf/2009/Dalton2009Pool.pdf},
timestamp = {2009.10.29}
}
Mobile sensors are an attractive proposition for environmental sensing, but pose significant engineering problems. Not least amongst these is the need to match the behaviour of the sensor platform to the physical environment in which it operates. We present initial work on using models of physical processes to generate models for autonomic control, and speculate that these can be used to improve the confidence we can place in sensed data.
@inproceedings{Dobson2009WellFounded,
author = {Dobson, Simon and Coyle, Lorcan and O'Hare, G.M.P. and Hinchey, Mike},
title = {From Physical Models to Well-Founded Control},
booktitle = {Sixth IEEE Conference and Workshops on Engineering of Autonomic and
Autonomous Systems},
year = {2009},
pages = {119-124},
owner = {lorcan},
pdf = {pdf/2009/Dobson2009WellFounded.pdf},
timestamp = {2009.09.28}
}
The area of activity identification is maturing well in the HCI and ubiquitous computing fields. However, although algorithm development is proceedings well, without publicly available datasets on which to compare results it is difficult to consolidate the disparate work being done. This problem exists because realistic datasets describing human activity are difficult and expensive to gather and because there are significant barriers to releasing the data once gathered. We review positive recent development with the release of two high-quality datasets. From our experiences using these datasets we list some recommendations for the gathering and release of future datasets. Finally, we propose a strategy of our own for gathering a new dataset from these recommendations.
@inproceedings{Coyle2009Gathering,
author = {Coyle, Lorcan and Ye, Juan and McKeever, Susan and Knox, Stephen and Stabeler, Matthew and Dobson, Simon and Nixon, Paddy},
title = {Gathering Datasets for Activity Identification},
booktitle = {Developing Shared Home Behavior Datasets to Advance HCI and Ubiquitous
Computing Research. Workshop at CHI'09, Boston, USA},
year = {2009},
owner = {lorcan},
pdf = {pdf/2009/Coyle2009Gathering.pdf},
timestamp = {2009.01.07}
}
Highly sensorised systems present two parallel challenges: how to design a sensor suite that can efficiently and cost-effectively support the needs of given services; and to extract the semantically relevant interpretations, or "situations", from the flood of context data collected by the sensors. We describe mathematical structures called situation lattices that can be used to address these two problems simultaneously, allowing designers to both design and refine situation identification whilst offering insights into the design of sensor suites. We validate the accuracy and efficiency of our technique against a third-party data set and demonstrate how it can be used to evaluate sensor suite designs.
@inproceedings{Ye09Percom,
author = {Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Using situation lattices in sensor analysis},
booktitle = {Proceedings of IEEE International Conference on Pervasive Computing
and Communications (PerCom 2009)},
year = {2009},
pages = {1-11},
month = mar,
dates = {9-13},
doi = {10.1109/PERCOM.2009.4912762},
isbn = {978-1-4244-3304-9},
pdf = {pdf/2009/Ye2009Using.pdf}
}
Through advances in sensing technology, a huge amount of data is available to context-aware applications. A major challenge is extracting features of this data that correlate to high-level human activities. Time, while being semantically rich and an essentially free source of information, has not received sufficient attention for this task. In this paper, we examine the potential for taking temporal features - inherent in human activities - into account when classifying them. Preliminary experiments using the PlaceLab dataset show that absolute time and temporal relationships between activities can improve the accuracy of activity classifiers.
@inproceedings{Ye2009Temporal,
author = {Ye, Juan and Clear, Adrian K. and Coyle, Lorcan and Dobson, Simon},
title = {On using temporal features to create more accurate human-activity
classifiers},
booktitle = {20th Conference on Artificial Intelligence and Cognitive Science,
UCD Dublin, Ireland},
year = {2009},
editor = {Coyle, Lorcan and Dunnion, John and Freyne, Jill},
pages = {274-283},
month = aug,
owner = {lorcan},
pdf = {pdf/2009/Ye2009Temporal.pdf},
timestamp = {2009.09.28}
}
In the domain of ubiquitous computing, the ability to identify the occurrence of situations is a core function of being ’context-aware’. Given the uncertain nature of sensor information and inference rules, reasoning techniques that cater for uncertainty hold promise for enabling the inference process. In our work, we apply the Dempster Shafer theory of evidence to infer situation occurrence with minimal use of training data. We describe a set of evidential operations for sensor mass functions using context quality and evidence accumulation for continuous situation detection. We demonstrate how our approach enables situation inference with uncertain information using a case study based on a published smart home activity data set.
@inproceedings{McKeever2009DempsterShafer,
author = {McKeever, Susan and Ye, Juan and Coyle, Lorcan and Dobson, Simon},
title = {Using Dempster-Shafer Theory of Evidence for Situation Inference},
booktitle = {4th European Conference on Smart Sensing and Context (EuroSSC)},
year = {2009},
pages = {149-162},
month = sep,
publisher = {Springer},
doi = {10.1007/978-3-642-04471-7_12},
owner = {lorcan},
pdf = {pdf/2009/McKeever2009DempsterShafer.pdf},
timestamp = {2009.10.20}
}
Much research on context quality in context-aware systems divides into two strands: (1) the qualitative identification of quality measures and (2) the use of uncertain reasoning techniques. In this paper, we combine these two strands, exploring the problem of how to identify and propagate quality through the different context layers in order to support the context reasoning process. We present a generalised, structured context quality model that supports aggregation of quality from sensor up to situation level. Our model supports reasoning processes that explicitly aggregate context quality, by enabling the identification and quantification of appropriate quality parameters. We demonstrate the efficacy of our model using an experimental sensor data set, gaining a significant improvement in situation recognition for our voting based reasoning algorithm.
@inproceedings{McKeever2009Context,
author = {McKeever, Susan and Ye, Juan and Coyle, Lorcan and Dobson, Simon},
title = {A Context Quality Model to Support Transparent Reasoning with Uncertain
Context},
booktitle = {1st International Workshop on Quality of Context (QuaCon), Stuttgart,
Germany},
year = {2009},
owner = {lorcan},
pdf = {pdf/2009/McKeever2009Context.pdf},
timestamp = {2009.09.28}
}
There are few areas of modern life in which software is not an important (though often invisible) component. The software in our lives is increasingly complex; its interaction with the real world means that its requirements are in a state of constant change (Lehman & Fernández-Ramil, 2006). Many non-software products and services, from healthcare to transport, education to business, depend on reliable, high-quality software. Software engineering is the activity that applies engineering principles to software. It applies systematic, rigorous discipline to the design and development of software, much as civil engineering does to construction. Software engineering improves the quality, reliability and predictability of software systems, by generating knowledge, tools and processes that both facilitate and improve the software development process. These qualities are essential wherever software failure might lead to significant safety, security, or economic losses. Software systems frequently need to be modified in response to changes in system requirements and in their operational environment (Swanson, 1976). Such modification may involve the addition of new functionality, the adjustment of existing functions, or the wholesale replacement of entire sub-systems. All such change is fraught with uncertainty — software projects involving change frequently fail to meet requirements, run over time and budget, or are abandoned (Rajlich and Bennett, 2000). As the ubiquity and complexity of software increase, a requirement has emerged for critical software which can successfully evolve without loss of quality — software that is engineered from the start to be easily changed, extended and reconfigured, while retaining its security, its performance, its reliability and predictability.
@techreport{Hinchey2009ECS,
author = {Hinchey, Mike and Coyle, Lorcan},
title = {Evolving Critical Systems},
institution = {Lero---the Irish Software Engineering Research Centre},
year = {2009},
number = {Lero-TR-2009-00},
address = {Lero---the Irish Software Engineering Research Centre, University
of Limerick, Ireland},
month = jul,
note = {Draft: 27 July 2009.},
owner = {lorcan},
pdf = {pdf/2009/Hinchey2009ECS.pdf},
timestamp = {2010.01.26},
url = {http://www.lero.ie/download.aspx?f=Lero-TR-2009-00-20090727.pdf}
}
Recent studies have shown that in distributed families keeping in touch is essential; this calls for technologies that can connect family members and bring them closer virtually. There are several social networking technologies online, but they are seldom designed for family connectedness and do not cater for the needs of computer-novice relatives. We present Near Dear, an application that brings online tools to an ambient display at home. The ambient display makes it easy for computer-novices to update and access online networking tools. We also conducted a user trial and evaluation of this system, which indicated that the developed system is convenient and intuitive.
@techreport{Hanumara2008Connecting,
author = {Hanumara, Poornima and Coyle, Lorcan},
title = {Connecting Families by Sharing the Minutiae of their Lives},
institution = {University College Dublin},
year = {2008},
number = {UCD-CSI-2008-05},
month = aug,
owner = {lorcan},
pdf = {pdf/2008/Hanumara2008Connecting.pdf},
timestamp = {2008.08.30}
}
While it is universally held by computer scientists that conference publications have a higher status in computer science than in other disciplines there is little quantitative evidence in support of this position. The importance of journal publications in academic promotion makes this a big issue since a focus on journal papers only will miss many significant papers published at conferences in computer science. In this paper we set out to quantify the relative importance of journal and conference papers in computer science. We show that computer science papers in leading conferences match the impact of papers in mid-ranking journals and surpass the impact of papers in journals in the bottom half of the ISI rankings - when impact is measured by citations in Google Scholar. We also show that there is a poor correlation between this measure of impact and conference acceptance rates. This indicates that conference publication is an inefficient market where venues that are equally challenging in terms of rejection rates offer quite different returns in terms of citations.
@techreport{Coyle2008Quantitative,
author = {Coyle, Lorcan and Freyne, Jill and Smyth, Barry and Cunningham, P\'adraig},
title = {A Quantitative Evaluation of the Relative Status of Journal and Conference
Publications in Computer Science},
institution = {University College Dublin},
year = {2008},
number = {UCD-CSI-2008-08},
month = oct,
owner = {lorcan},
pdf = {pdf/2008/Coyle2008Quantitative.pdf},
timestamp = {2008.10.07},
url = {http://www.csi.ucd.ie/content/quantitative-evaluation-relative-status-journal-and-conference-publications-computer-science}
}
@inproceedings{Dobson2008Open,
author = {Dobson, Simon and Stevenson, Graeme and Williamson, Graham and Knox, Stephen and Stabeler, Matthew and Coyle, Lorcan and Neely, Steve and Nixon, Paddy},
title = {An Open-Source Infrastructure for Pervasive Computing},
booktitle = {PerAda Magazine},
year = {2008},
month = oct,
doi = {10.1117/2.1200809.1262},
owner = {lorcan},
pdf = {pdf/2008/Dobson2008Open.pdf},
timestamp = {2008.10.07},
url = {http://www.perada-magazine.eu/view.php?article=1262-2008-09-22&category=Middleware}
}
Presence is an important part of our day to day lives. Often we will have a sense of who is around us and what they are doing by the sounds of doors closing, cupboards banging, footsteps on floors, voices vaguely heard through walls, etc. In digital spaces, such as GUI desktops, presence enhances our sense of connection with geographical separate friends and colleagues. In this paper we report on Ambient Jewelry, which is a project exploring the intersection of individual and user generated customization with ambient presence displays. With this research we are seeking techniques that enable people to invent, discover and find new forms of ambient presence visualisations.
@inproceedings{Lado2008User,
author = {Insua, Germ\'an Lado and Bennett, Mike and Nixon, Paddy and Coyle, Lorcan},
title = {User Generated Ambient Presence},
booktitle = {Proc. of 2nd Workshop on Ambient Information Systems. Colocated with
Ubicomp 2008, Seoul, South Korea.},
year = {2008},
editor = {Hazlewood, William R. and Coyle, Lorcan and Pousman, Zachary and Lim, Youn-kyung},
volume = {402},
series = {CEUR Workshop Proceedings ISSN 1613-0073},
month = sep,
pdf = {pdf/2008/Lado2008User.pdf},
url = {http://CEUR-WS.org/Vol-402/paper05.pdf}
}
Applications that rely on mobile devices for user interaction must be mindful of the user’s limited attention, which will typically be split between several competing tasks. Content delivery in such systems must be adapted closely to users’ evolving situations and shifting priorities, in a way that cannot be accomplished using static filtering determined a priori. We propose a more dynamic context-driven approach to content delivery. We demonstrate our approach using Scatterbox, a pervasive computing application we have developed which performs sensor fusion to derive a user’s current situation. Based on the user’s level of interruptibility, Scatterbox prioritises and forwards relevant messages to their mobile phone. We draw conclusions from a preliminary evaluation of the system.
@article{Knox2008Scatterbox,
author = {Knox, Stephen and Shannon, Ross and Coyle, Lorcan and Clear, Adrian and Dobson, Simon and Quigley, Aaron and Nixon, Paddy},
title = {Scatterbox: Context-Aware Message Management},
journal = {Revue d'Intelligence Artificielle},
year = {2008},
volume = {22},
pages = {549-568},
number = {5},
doi = {10.3166/ria.22.549-568},
owner = {lorcan},
pdf = {pdf/2008/Knox2008Scatterbox},
timestamp = {2008.11.04}
}
Context-aware systems typically use data sensed from the environment to drive adaptive behaviour. Sensed data is inherently imprecise and uncertain; in addition, new uncertainties are introduced when sensed data is fused with other data to infer context at the more abstract level of situations. We present an uncertainty model which aggregates context uncertainty and provides mechanisms to capture uncertainty at situation level. We demonstrate the application of the model in a sample scenario using an experimental data set.
@inproceedings{Mckeever2008Multilayered,
author = {McKeever, Susan and Ye, Juan and Coyle, Lorcan and Dobson, Simon},
title = {A Multilayered Uncertainty Model for Context Aware Systems},
booktitle = {Late Breaking Results - Adjunct Proceedings of the 6th International
Conference on Pervasive Computing},
year = {2008},
pages = {1-4},
address = {Sydney, Australia},
month = may,
publisher = {OCG},
owner = {lorcan},
pdf = {pdf/2008/Mckeever2008Multilayered.pdf},
timestamp = {2008.11.04}
}
Stock photo libraries are the most common means for publishers and advertisers to find images for their media. Searching for the perfect photo can be a time-consuming and frustrating task. This is because searching is often dependent on the descriptors or tags given to each photo by the editors and contributors to the library. The tagging process is subjective, further complicating the search process. We describe an algorithm that uses domain ontologies to improve the interactions with these libraries. Ontologies are used to expand query terms based on users’ initial search queries. We present results that demonstrate that the use of ontologies greatly improves users ability to retrieve photos when undertaking a number of search tasks.
@inproceedings{Murdoch2008Ontology,
author = {Murdoch, Olga and Coyle, Lorcan and Dobson, Simon},
title = {Ontology-Based Query Recommendation as a Support to Image Retrieval},
booktitle = {19th Irish Conference on Artificial Intelligence and Cognitive Science,
Cork, Ireland},
year = {2008},
editor = {Derek Bridge, Ken Brown, Barry O'Sullivan and Sorensen, Humphrey},
pages = {103-112},
month = aug,
owner = {lorcan},
pdf = {pdf/2008/Murdoch2008Ontology.pdf},
timestamp = {2008.08.30}
}
It is becoming difficult to convey information from an ever-increasing number of digital sources to users in a condensed and meaningful way. This growth has particularly occurred with peripheral information sources. These are of general interest to users, but do no require or typically command constant focus or attention. Examples include weather, stock data, blogs, and calendars. Ambient Displays present information unobtrusively in an intelligent fashion using abstract visual cues and metaphors and have the possibility of acting as a complement to information filtering systems. We describe the implementation of an ambient display that contains elements representing time, weather, public transport departure times, and the proximity of friends. An initial impact study was undertaken and found a high sense of usefulness and curiosity in the finished application and in the field as a whole.
@inproceedings{Phelan2008Ambient,
author = {Phelan, Owen and Coyle, Lorcan and Stevenson, Graeme and Neely, Steve},
title = {The Ambient Calendar},
booktitle = {19th Irish Conference on Artificial Intelligence and Cognitive Science,
Cork, Ireland},
year = {2008},
editor = {Derek Bridge, Ken Brown, Barry O'Sullivan and Sorensen, Humphrey},
pages = {282-290},
owner = {lorcan},
pdf = {pdf/2008/Phelan2008Ambient.pdf},
timestamp = {2008.08.30}
}
Situations, the semantic interpretations of context, provide a better basis for selecting adaptive behaviours than context itself. The definition of situations typically rests on the abilityto define logical expressions and inference methods to identify particular situations. In this paper we extend this approach to provide for efficient organisation and selection in systems with large numbers of situations having structured relationships to each other. We apply lattice theory to define a specialisation relationship across situations, and show how this can be used to improve the identification of situations using lattice operators and uncertain reasoning. We demonstrate the technique against a real-world dataset.
@article{Ye2008Representing,
author = {Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Representing and Manipulating Situation Hierarchies using Situation
Lattices},
journal = {Revue d'Intelligence Artificielle},
year = {2008},
volume = {22},
pages = {647-667},
number = {5},
doi = {10.3166/ria.22.647-667},
owner = {lorcan},
pdf = {pdf/2008/Ye2008Representing},
timestamp = {2008.11.04}
}
Pervasive computing is typically highly sensor-driven, but sensors provide only evidence of fact rather than facts themselves. The uncertainty of sensor data will affect each component in a pervasive computing system, which may decrease the quality of its provided services. We provide a general model to represent semantics of uncertainty in different levels (e.g., sensor, lower-level context and higher-level context). Within our model, fine-grained approaches are applied to evaluate and propagate uncertainties. They will help to resolve the uncertainty in each process of context management so that the effect of uncertainty on system services will be minimised.
@inproceedings{Ye2008Resolving,
author = {Ye, Juan and McKeever, Susan and Coyle, Lorcan and Neely, Steve and Dobson, Simon},
title = {Resolving uncertainty in context integration and abstraction: context
integration and abstraction},
booktitle = {ICPS '08: Proceedings of the 5th international conference on Pervasive
services},
year = {2008},
pages = {131-140},
address = {New York, NY, USA},
publisher = {ACM},
doi = {http://doi.acm.org/10.1145/1387269.1387292},
isbn = {978-1-60558-135-4},
location = {Sorrento, Italy},
pdf = {pdf/2008/Ye2008Resolving}
}
Context-aware systems that rely on mobile devices for user interaction must address the low bandwidth of both communications and more importantly the user’s limited attention, which will typically be split between several competing tasks. Content delivery in such systems must be adapted closely to users’ evolving situations and shifting priorities, in a way that cannot be accomplished using static filtering determined a priori. We propose a more dynamic context-driven approach to content delivery, that integrates information from a wide range of sources. We demonstrate our approach on a system for adaptive message prioritisation and forwarding.
@inproceedings{Knox2007Towards,
author = {Knox, Stephen and Clear, Adrian K. and Shannon, Ross and Coyle, Lorcan and Dobson, Simon and Quigley, Aaron and Nixon, Paddy},
title = {Towards Scatterbox: a Context-Aware Message Forwarding Platform},
booktitle = {Fourth International Workshop on Modeling and Reasoning in Context
(MRC 2007)},
year = {2007},
pages = {13--24},
month = aug,
pdf = {pdf/2007/Knox2007Towards.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1183397642350.pdf}
}
Pervasive computing is by its nature open and extensible, and must integrate information from a diverse range of sources. This leads to a problem of information exchange, so sub-systems must agree on shared representations. Ontologies potentially provide a well-founded mechanism for the representation and exchange of such structured information. A number of ontologies have been developed specifically for use in pervasive computing, none of which appears to cover adequately the space of concerns applicable to application designers. We compare and contrast the most popular ontologies, evaluating them against the system challenges generally recognised within the pervasive computing community. We identify a number of deficiencies that must be addressed in order to apply ontological techniques successfully to next-generation pervasive systems.
@article{Ye2007Ontology,
author = {Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Ontology-based Models in Pervasive Computing Systems},
journal = {The Knowledge Engineering Review},
year = {2007},
volume = {22},
pages = {315--347},
month = dec,
doi = {10.1017/S0269888907001208},
eprint = {http://journals.cambridge.org/article_S0269888907001208},
issue = {04},
pdf = {http://www.csi.ucd.ie/UserFiles/publications/1197995762823.pdf}
}
With the general increase of life span that our advances in health care have afforded us, more people are suffering from short term memory loss than ever before. Short term memory sufferers often forget what they were doing in the middle of a task and can find themselves in dangerous situations, such as leaving the stove on and leaving the house. They could benefit from an RFID based reminder system that would determine what they were doing based on what objects they touch. To use the system, the user wears an RFID glove which has a reader in the palm. The RFID glove reads the tags on the nearby objects. Along with the RFID glove we are developing an application that enables the user to interact with a reminder application. The application alerts the user of important activities they may have forgotten they started and when an activity is interrupted. It also keeps a record of the list of activities they have performed and objects they have touched through out the day.
@techreport{Lustig2007Reminding,
author = {Lustig, Caitlin and Coyle, Lorcan},
title = {Reminding Short-Term Memory Sufferers to Complete Routine Tasks},
institution = {University College Dublin},
year = {2007},
number = {UCDCSI Technical Report 2007-10},
month = aug,
pdf = {pdf/2007/Lustig2007Reminding.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/UCD-CSI-2007-10.pdf}
}
The progress of health care in the western world has been marked by an increase in life expectancy. Advances in life expectancy have meant that more people are living with acute health problems, many of which are related to impairment of memory. This paper describes a pair of scenarios that use RFID to assist people who may suffer from memory defects to extend their capability for independent living. We present our implementation of an RFID glove, describe its operation, and show how it enables the application scenarios.
@inproceedings{Lustig2007Using,
author = {Lustig, Caitlin and Novatchkov, Hristo and Dunne, Lucy and McHugh, Mike and Coyle, Lorcan},
title = {Using Colocation to Support Human Memory},
booktitle = {MeMos 2007: Supporting Human Memory with Interactive Systems. Workshop
at the 2007 British HCI International Conference.},
year = {2007},
pages = {41--44},
month = sep,
pdf = {pdf/2007/Lustig2007Using.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1188303216163.pdf}
}
Construct differs from other pervasive systems platforms in a number of key respects. It is completely standards-based, using RDF as its data exchange model and ZeroConf for resource discovery. It supports a knowledge-centric model of interaction where clients’ actions are driven by queries and triggers about the context of the system. It uses gossiping to maintain a consistent state across a distributed data structure, which maximises robustness and scalability and avoids many problems with hot-spots and hot-paths in communications. Finally, it treats all information sources uniformly as sensors acting as inputs to uncertain reasoning algorithms.
@inproceedings{Dobson2007Construct,
author = {Dobson, Simon and Nixon, Paddy and Coyle, Lorcan and Neely, Steve and Stevenson, Graeme and Williamson, Graham},
title = {Construct: An Open Source Pervasive Systems Platform},
booktitle = {Consumer Communications and Networking Conference (CCNC), 2007, 4th
IEEE, Las Vegas, NV, USA.},
year = {2007},
pages = {1203-1204},
month = jan,
isbn = {1-4244-0667-6},
pdf = {pdf/2007/Dobson2007Construct.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1188312808293.pdf}
}
Smart homes are sensor-rich environments that contain dynamic sets of interacting components. These components often use competing and closed standards and form a message-based architecture. This complicates the development of applications that require information from disparate sources. It becomes difficult to add new components or to allow components from different applications to interact with each another. In this paper we describe Construct, a pervasive computing middleware that is ideally suited for deployment in the smart home. Construct acts as a sensor fusion layer that takes output from each smart home component and makes it available to all applications. This makes it easy to develop applications that require access to heterogeneous sources of sensor data, and to add sensors to existing systems to improve their performance. This paper demonstrates two Construct-enabled smart home applications and shows how access to new sensors leads to improvements in their performance.
@article{Coyle2007Sensor,
author = {Coyle, Lorcan and Neely, Steve and Stevenson, Graeme and Sullivan, Mark and Dobson, Simon and Nixon, Paddy},
title = {Sensor Fusion-Based Middleware for Smart Homes},
journal = {International Journal of Assistive Robotics and Mechatronics (IJARM)},
year = {2007},
volume = {8},
pages = {53-60},
number = {2},
month = jun,
issn = {1975-0153},
pdf = {pdf/2007/Coyle2007Sensor.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1178110026850.pdf}
}
Location detection systems that use tags are a popular means of determining a user’s location. These systems are characterised as requiring the user to carry an identity tag that is detected by sensors, which typically use some form of triangulation to determine location. Although estimates for precision for these systems are published by the respective manufacturers the customer experience can vary widely. This paper proposes an evaluation framework which will allow different systems to be compared more directly. This framework is specifically targeted at evaluating the experiences of tagging humans, which can cause particular difficulties due to the fact that many tag-based systems use communication frequencies that cannot pass easily through the human body.
@inproceedings{Coyle2007Proposed,
author = {Coyle, Lorcan and Ye, Juan and Loureiro, Emerson and Knox, Stephen and Dobson, Simon and Nixon, Paddy},
title = {A Proposed Approach to Evaluate the Accuracy of Tag-based Location
Systems},
booktitle = {In USE 07: Workshop on Ubiquitous Systems Evaluation, Innsbruck,
Austria, 2007. UbiComp 2007 Workshop Proceedings},
year = {2007},
editor = {Neely, Steve and Stevenson, Graeme and Terzis, Sotirios},
pages = {292--296},
month = sep,
pdf = {pdf/2007/Coyle2007Proposed.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1190630881297.pdf}
}
Location-aware systems provide customised services or applications according to users’ locations. While much research has been carried out in developing models to represent location information and spatial relationships, it is usually limited to modelling simple environments (cf. [13,19,3]). This paper proposes a unified space model for more complex environments (e.g., city plan or forest). This space model provides a flexible, expressive, and powerful spatial representation. It also proposes a new data structure - an integrated lattice and graph model - to express comprehensive spatial relationships. This structure not only provides multiple graphs at different abstraction levels, but it also collapses the whole map into smaller local graphs. This mechanism is beneficial in reducing the complexity of creating and maintaining a map and improving the efficiency of path finding algorithms.
@inproceedings{Ye2007Unified,
author = {Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {A Unified Semantics Space Model},
booktitle = {Location- and Context-Awareness},
year = {2007},
editor = {Hightower, Jeffrey and Schiele, Bernt and Strang, Thomas},
volume = {4718},
series = {LNCS},
pages = {103--120},
publisher = {Springer},
isbn = {978-3-540-75159-5},
location = {Heidelberg},
pdf = {pdf/2007/Ye2007Unified.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1190628642708.pdf}
}
Conversational recommender systems have recently emerged as useful alternative strategies to their single-shot counterpart, especially given their ability to expose a user’s current preferences. These systems use conversational feedback to hone in on the most suitable item for recommendation by improving the mechanism that finds useful collaborators. We propose a novel architecture for performing recommendation that incorporates information about the individual performance of neighbours during a recommendation session, into the neighbour retrieval mechanism. We present our architecture and a set of preliminary evaluation results that suggest there is some merit to our approach. We examine these results and discuss what they mean for future research.
@inproceedings{Rafter2007Sticking,
author = {Rafter, Rachael and Coyle, Lorcan and Nixon, Paddy and Smyth, Barry},
title = {Sticking with a Winning Team: Better Neighbour Selection for Conversational
Collaborative Recommendation},
booktitle = {Proceedings of the 18th Irish Conference on Artificial Intelligence
and Cognitive Science, 29th-31st August 2007},
year = {2007},
month = aug,
pdf = {pdf/2007/Rafter2007Sticking.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1187096214235.pdf}
}
Context awareness in Case-Based Reasoning (CBR) systems has become a topic of increased research of late. In CBR, context serves as a major source for reasoning, decision-making, and adaptation. Achieving context-awareness for context-sensitive CBR systems will depend on their ability to represent and manipulate in formation about a rich range of contextual factors. These factors may include not only physical characteristics of the task environment, but many other aspects such as the knowledge states (of both the application and user), and user beliefs and emotions. There presentation and reasoning problem therein presents research challenges to which numerous methods and techniques derived from artificial intelligence and knowledge management (e.g., logical reasoning, object relationship models, ontologies, similarity measures, and intelligent retrieval mechanisms) are now being brought to bear. This workshop served as a discussion platform to researchers and practitioners exploring issues and approaches for context-sensitive systems involving CBR to share their problems and techniques. The discussion extended towards mechanisms and techniques for structured storage of contextual information, effective ways to retrieve, reuse, and adapt it, as well as methods for enabling integration of context and application knowledge. The main question raised at the workshop is how to deal with contextual and/or contextualized in formation, e.g., contextualized cases for a CBR system. To kickstart this discussion, three selected papers we represented, which opened the discussions with specific questions about context-awareness, explanations, and context ontology issues.
@proceedings{Coyle2007Case,
title = {2nd Workshop on Case Based Reasoning and Context-Awareness},
year = {2007},
editor = {Coyle, Lorcan and Schwarz, Sven},
volume = {254},
series = {CEUR Workshop Proceedings ISSN 1613-0073},
month = aug,
url = {http://CEUR-WS.org/Vol-271}
}
The concept of calm technology, introduced by Mark Weiser, has led researchers from several disciplines to explore new and unconventional ways of conveying information. Some well-known examples of such novel information techniques include Ambient Devices’ Stock Orb, Koert van Mensvoort’s Datafountain, Violet’s Nabaztag, Jafarinami et al.’s Breakaway, Mynatt et al.’s Audio Aura and Digital Family Portrait, Mankoff et al.’s Daylight Display and BusMobile, and Natalie Jeremijenko’s Dangling String. Applications such as these, that publish information in a non-intrusive or calm manner are Ambient Information Systems. The 1st International Workshop on the Design and Evaluation of Ambient Information Systems was held in Toronto, Canada, on May 13th 2007, in conjunction with the 5th International Conference on Pervasive Computing. The goal of this workshop was to bring researchers together to discuss this domain of growing interest in both pervasive computing and human-computer interaction. This domain is described by mechanisms that are minimally attended and perceivable from outside the range of a person’s direct attention, providing pre-attentive processing without being overly distracting. Developing new technologies such as these poses new and difficult challenges. These technologies display information outside of a person’s direct attention, which is a space that is not currently well understood, making it difficult to evaluate their effectiveness. A great deal of care is required to design studies which accurately observe the effect of ambient devices, particularly since the test subjects are not meant to observe them directly. After all, how does one provide a subject with a device and say, "I want you to use this, but please do not think about it?" Our workshop sought to gather perspectives from researchers in the field on these and other problems. Eleven excellent submissions were accepted which describe works-in-progress, frameworks, taxonomies, methodologies, evaluation paradigms, and case studies, and are published here.
@proceedings{Hazlewood2007Ambient,
title = {1st Workshop on Ambient Information Systems},
year = {2007},
editor = {Hazlewood, William R. and Coyle, Lorcan and Consolvo, Sunny},
volume = {254},
series = {CEUR Workshop Proceedings ISSN 1613-0073},
month = may,
pdf = {pdf/2007/Hazlewood2007Ambient.pdf},
url = {http://CEUR-WS.org/Vol-254/proceedings.pdf}
}
Much recent research has focused on using situations rather than individual pieces of context as a means to trigger adaptive system behaviour. While current research on situations emphasises their representation and composition, they do not provide an approach on how to organise and identify their occurrences efficiently. This paper describes how lattice theory can be utilised to organise situations, which reflects the internal structure of situations such as generalisation and dependence. We claim that situation lattices will prove beneficial in identifying situations, and maintaining the consistency and integrity of situations. They will also help in resolving the uncertainty issues inherent in context and situations by working with Bayesian Networks.
@inproceedings{Ye2007Using,
author = {Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Using Situation Lattices to Model and Reason about Context},
booktitle = {Fourth International Workshop on Modeling and Reasoning in Context
(MRC 2007).},
year = {2007},
pages = {1--12},
month = aug,
pdf = {pdf/2007/Ye2007Using.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1183397975555.pdf}
}
The vision of pervasive computing is that objects, buildings and environments may be endowed with software intelligence to improve human interactions both with the individual objects and with the system as a whole. Realising this dream is posing significant challenges for designers - allowing individual applications to co-exist in a common space without interfering, making the capabilities of new sensors and services available to other applications as they appear and providing the necessary decentralised control to obtain robust behaviour.
@misc{Dobson2006Towards,
author = {Dobson, Simon and Neely, Steve and Stevenson, Graeme and Coyle, Lorcan and Nixon, Paddy},
title = {Towards a platform for widespread embedded intelligence},
howpublished = {ERCIM News 67. Special Theme: Embedded Intelligence},
month = oct,
year = {2006},
owner = {lorcan},
timestamp = {2007.11.13},
url = {http://ercim-news.ercim.org/content/view/41/54/}
}
Systems for home automation can make a vital contribution to the well-being of individuals requiring moderate amounts of support for day-to-day living. Existing systems suffer both from competing and often closed standards bases and from a message-based architecture that can complicate the development of flexible applications requiring information from disparate sources. We describe a knowledge-based pervasive computing middleware and show how it can be used to provide semantically rich unification over a range of home- and web-based automation systems.
@inproceedings{Coyle2006SensorA,
author = {Coyle, Lorcan and Neely, Steve and Rey, Ga{\"e}tan and Stevenson, Graeme and Sullivan, Mark and Dobson, Simon and Nixon, Paddy},
title = {Sensor Fusion-Based Middleware for Assisted Living},
booktitle = {Proc. of 1st International Conference On Smart homes \& heath Telematics
(ICOST'2006) "Smart Homes and Beyond" },
year = {2006},
pages = {281--288},
publisher = {IOS Press},
city = {Belfast, UK},
pdf = {pdf/2006/Coyle2006Sensor.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1145369803142.pdf}
}
The dissemination of context data across a pervasive environment has proven to be a difficult problem. Techniques using gossiping algorithms offer simplicity and flexibility but often result in poor performance with respect to timeliness of delivery and communication cost. In this ongoing work, we present enhanced gossiping algorithms that aim to improve the efficiency of context data delivery in a decentralised manner using network and data-driven approaches.
@inproceedings{Anawar2006Context,
author = {Anawar, Syarulnaziah and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Context Delivery in Ad Hoc Networks Using Enhanced Gossiping Algorithms.},
booktitle = {EuroSSC},
year = {2006},
pages = {218--221},
pdf = {pdf/2006/Anawar2006Context.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1155980200071.pdf}
}
Complex and dynamic working environments such as healthcare facilities consist of staff, patients and equipment constantly moving in response to changing medical requirements. Knowing the current location of people and equipment is essential for the smooth running of a facility, yet creating a global view through tracking is a challenging task. It is clear that many common hospital situations can be improved with real-time access to the various actors’ location information. One of the main problems with implementing such services is that current location based applications tend to be proprietary and the data generated closed. The realisation of ubiquitous location based services demands the exploration of hybrid models and methods that can utilise existing and subsequent infrastructures in novel and complimentary ways. We describe a number of hospital scenarios that use location-based services and make available all the location data gathered. We propose that by aggregating location data by a range of acquisition methods it is possible to improve the performance of location applications and readily adapt to the introduction of new location detection technologies.
@inproceedings{Coyle2006SensorB,
author = {Coyle, Lorcan and Neely, Steve and Nixon, Paddy and Quigley, Aaron},
title = {Sensor Aggregation and Integration in Healthcare Location Based Services},
booktitle = {1st Workshop on Location Based Services for Health Care},
year = {2006},
pages = {1-4},
month = nov,
doi = {10.1109/PCTHEALTH.2006.361698},
pdf = {pdf/2006/Coyle2006SensorB.pdf},
url = {http://www.cs.ucd.ie/UserFiles/publications/1160061572429.pdf}
}
Group recommender systems introduce a whole set of new challenges for recommender systems research. The notion of generating a set of recommendations that will satisfy a group of users, with potentially competing interests, is challenging in itself. In addition to this we must consider how to record and combine the preferences of many different users as they engage in simultaneous recommendation dialogs. In this paper we introduce a group recommender system that is designed to provide assistance to a group of friends trying to plan a skiing vacation.
@inproceedings{Mccarthy2006Group,
author = {McCarthy, Kevin and Salam{\'o}, Maria and Coyle, Lorcan and McGinty, Lorraine and Smyth, Barry and Nixon, Paddy},
title = {Group Recommender Systems: A Critiquing Based Approach},
booktitle = {Proceedings of the 11th International Conference on Intelligent User
Interfaces (IUI '06), ACM Press, Sydney, Australia.},
year = {2006},
editor = {Paris, C. and Sidner, C.},
pages = {267-269},
pdf = {pdf/2006/Mccarthy2006Group.pdf},
url = {http://www.cs.ucd.ie/UserFiles/publications/1141386468378.pdf}
}
Group recommender systems introduce a whole set of new challenges for recommender systems research. The notion of generating a set of recommendations that will satisfy a group of users with potentially competing interests is challenging in itself. In addition to this we must consider how to record and combine the preferences of many different users as they engage in simultaneous recommendation dialogs. In this paper we introduce a group recommender system that is designed to provide assistance to a group of friends trying the plan a skiing vacation. The system uses the DiamondTouch interactive tabletop to allow up to 4 users to simultaneously engage in parallel recommendation sessions and we describe how personal and shared profiles and interaction spaces can be managed to generate sets of recommendations for the individual as well as the group.
@inproceedings{Mccarthy2006Cats,
author = {McCarthy, Kevin and Salam{\'o}, Maria and Coyle, Lorcan and McGinty, Lorraine and Smyth, Barry and Nixon, Paddy},
title = {CATS: A Synchronous Approach to Collaborative Group Recommendation},
booktitle = {Proceedings of the Nineteenth International Florida Artificial Intelligence
Research Society Conference, Melbourne Beach, Florida, USA, May 11-13,
2006},
year = {2006},
editor = {Sutcliffe, Geoff and Goebel, Randy},
pages = {86-91},
month = may,
publisher = {AAAI Press}
}
We propose that traditional case-based recommender systems can be improved by informing them with context data describing the user’s environment. We outline existing applications with similar objectives and describe an application of our own - Ticketyboo - which uses music listening preferences and context information from users’ calendars to recommend tickets for music concerts. This data is gathered by virtual sensors that monitor each user’s music player and calendar applications. The novelty of this approach is that context data is provided to Ticketyboo via a dedicated context infrastructure. This results in a clear separation between the providers and consumers of context data. By utilising context data in this way, minimal user input/feedback is required to guide the system since the need for explicit user feedback is negated.
@inproceedings{Coyle2006Supplementing,
author = {Coyle, Lorcan and Balfe, Evelyn and Stevenson, Graeme and Neely, Steve and Dobson, Simon and Nixon, Paddy and Smyth, Barry},
title = {Supplementing Case-based Recommenders with Context Data},
booktitle = {Proceedings of the 1st Workshop on Case-Based Reasoning and Context
Awareness at the 8th European Conference on Case-Based Reasoning,
Turkey, September 5, 2006, CEUR Workshop Proceedings, ISSN 1613-0073,
online CEUR-WS.org/Vol-221/03.pdf.},
year = {2006},
pdf = {pdf/2006/Coyle2006Supplementing.pdf},
url = {http://ceur-ws.org/Vol-221/03.pdf}
}
There is a commonly accepted need for contexts and ontologies to describe the vast amounts of data that are available to pervasive computing applications. Existing contexts and ontologies are either much generalised, very application specific, or inflexible. An integrated approach is required in which new concepts can be added and related to existing ones transparently. This paper describes a novel approach to the design of a set of contexts and ontologies for context-aware pervasive computing systems. It describes a Query Service, that lies between applications and contextual information, which complemented by the contexts and ontologies, offers a more powerful query answering service to application developers than is currently available.
@inproceedings{Clear2006Integrating,
author = {Clear, Adrian K. and Knox, Stephen and Ye, Juan and Coyle, Lorcan and Dobson, Simon and Nixon, Paddy},
title = {Integrating Multiple Contexts and Ontologies in a Pervasive Computing
Framework},
booktitle = {Contexts and Ontologies: Theory, Practice and Applications},
year = {2006},
pages = {20-25},
address = {Riva Del Garda, Italy},
month = aug,
pdf = {pdf/2006/Clear2006Integrating.pdf},
url = {http://CEUR-WS.org/Vol-210/paper5.pdf}
}
Pervasive computing systems require large amounts of information to be available to devices in order to support context-aware applications. Information must be routed from the sensors that provide it to the applications that consume it in a timely fashion. However, the potential size and ad hoc nature of these environments makes the management of communications a non-trivial task. One proposed solution to this problem uses gossiping, a class of probabilistic routing protocol, to disseminate context information throughout the environment. Gossiping algorithms require far less in the way of guarantees about network structure, reliability, and latency than alternative approaches, but are unproven in real world scenarios. We describe the on-going development of a framework for evaluating the performance of these algorithms within the context of pervasive environments.
@inproceedings{Williamson2006Scalable,
author = {Williamson, Graham and Stevenson, Graeme and Neely, Steve and Coyle, Lorcan and Nixon, Paddy},
title = {Scalable information dissemination for pervasive systems: implementation
and evaluation},
booktitle = {MPAC '06: Proceedings of the 4th international workshop on Middleware
for Pervasive and Ad-Hoc Computing (MPAC 2006)},
year = {2006},
pages = {7-13},
address = {New York, NY, USA},
publisher = {ACM Press},
isbn = {1-59593-421-9},
location = {Melbourne, Australia},
pdf = {pdf/2006/Williamson2006Scalable.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1160061136893.pdf}
}
Much work has been done in the areas of distributed and shared interface environments and intelligent and context aware applications. This paper describes our goals in the enhancement of a collaborative travel recommender deployed with a DiamondTouch interface to encompasses these technologies. The enhancements we propose will allow users to interact with installations on remote DiamondTouch devices and will grant the application access to an abundance of context data.
@inproceedings{Coyle2006Developing,
author = {Coyle, Lorcan and Rey, Ga{\"e}tan and McCarthy, Kevin and Salam{\'o}, Maria and McGinty, Lorraine and Smyth, Barry and Nixon, Paddy},
title = {Developing a Distributed Context-Aware Collaborative Recommender
System},
booktitle = {Proceedings of the Workshop on Multi-User and Ubiquitous User Interfaces
at the 10th International Conference on Intelligent User Interfaces
(IUI-06), Sydney, Australia.},
year = {2006},
editor = {Buth, A. and Kray, C. and Kruger, A. and Schwesig, C.},
pages = {8-9},
pdf = {pdf/2006/Coyle2006Developing.pdf},
url = {http://www.cs.ucd.ie/UserFiles/publications/1142348764744.pdf}
}
Event-based systems are a popular substrate for distributing information derived from sensors to be used in driving adaptive behaviour. This paper argues that using events directly provides a poor model of context, and that a hybrid approach that uses events to populate and maintain a distributed knowledge base offers a more stable solution. The inherent uncertainties in both sensor data and reasoning imply that traditional knowledge-based system techniques applied to context be extended to deal with more uncertain reasoning
@article{Dobson2006Hybridising,
author = {Dobson, Simon and Coyle, Lorcan and Nixon, Paddy},
title = {Hybridising events and knowledge as a basis for building autonomic
systems},
journal = {Journal of Trusted and Autonomic Computing},
year = {2006},
pdf = {pdf/2006/Dobson2006Hybridising.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1145365322729.pdf}
}
Spam filtering is a particularly challenging machine learning task as the data distribution and concept being learned changes over time. It exhibits a particularly awkward form of concept drift as the change is driven by spammers wishing to circumvent spam filters. In this paper we show that lazy learning techniques are appropriate for such dynamically changing contexts. We present a case-based system for spam filtering that can learn dynamically. We evaluate its performance as the case-base is updated with new cases. We also explore the benefit of periodically redoing the feature selection process to bring new features into play. Our evaluation shows that these two levels of model update are effective in tracking concept drift.
@article{Delany2005Case,
author = {Delany, Sarah Jane and Cunningham, P{\'a}draig and Tsymbal, Alexey and Coyle, Lorcan},
title = {A Case-Based Technique for Tracking Concept Drift in Spam Filtering},
journal = {Knowledge-Based Systems},
year = {2005},
volume = {18},
pages = {187--195},
number = {4--5},
pdf = {pdf/2005/Delany2005Case.pdf},
publisher = {Elsevier},
url = {http://www.csi.ucd.ie/UserFiles/publications/1146565915422.pdf}
}
Because of the changing nature of spam, a spam filtering system that uses machine learning will need to be dynamic. This suggests that a case-based (memory-based) approach may work well. Case-Based Reasoning (CBR) is a lazy approach to machine learning where induction is delayed to run time. This means that the case base can be updated continuously and new training data is immediately available to the induction process. In this paper we present a detailed description of such a system called ECUE and evaluate design decisions concerning the case representation. We compare its performance with an alternative system that uses Naı̈ve Bayes. We find that there is little to choose between the two alternatives in cross-validation tests on data sets. However, ECUE does appear to have some advantages in tracking concept drift over time.
@article{Delany2005Assessment,
author = {Delany, Sarah Jane and Cunningham, P{\'a}draig and Coyle, Lorcan},
title = {An Assessment of Case-Based Reasoning for Spam Filtering},
journal = {Artificial Intelligence Review},
year = {2005},
volume = {24},
pages = {359--378},
number = {3--4},
pdf = {pdf/2005/Delany2005Assessment.pdf},
publisher = {Springer Science+Business Media B.V.},
url = {http://www.csi.ucd.ie/UserFiles/publications/1146565959912.pdf}
}
In this paper we describe ConStruct, a distributed, decentralised infrastructure for the collection, processing and distribution of context information in a ubiquitous computing environment.
@inproceedings{Stevenson2005Construct,
author = {Stevenson, Graeme and Coyle, Lorcan and Neely, Steve and Dobson, Simon and Nixon, Paddy},
title = {ConStruct --- A Decentralised Context Infrastructure for Ubiquitous
Computing Environments},
booktitle = {IT\&T Annual Conference, Cork Institute of Technology, Ireland},
year = {2005},
pdf = {pdf/2005/Stevenson2005Construct.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1145364733079.pdf}
}
Clearly, machine learning techniques can play an important role in filtering spam email because ample training data is available to build a robust classifier. However, spam filtering is a particularly challenging task as the data distribution and concept being learned changes over time. This is a particularly awkward form of concept drift as the change is driven by spammers wishing to circumvent the spam filters. In this paper we show that lazy learning techniques are appropriate for such dynamically changing contexts. We present a case-based system for spam filtering called ECUE that can learn dynamically. We evaluate its performance as the case-base is updated with new cases. We also explore the benefit of periodically redoing the feature selection process to bring new features into play. Our evaluation shows that these two levels of model update are effective in tracking concept drift.
@inproceedings{Delany2004Case,
author = {Delany, Sarah Jane and Cunningham, P{\'a}draig and Tsymbal, Alexey and Coyle, Lorcan},
title = {A Case-Based Technique for Tracking Concept Drift in Spam Filtering},
booktitle = {Applications and Innovations in Intelligent Systems XII, Procs. of
AI 2004},
year = {2004},
editor = {Macintosh, A. and Ellis, R. and Allen, T.},
pages = {3--16},
publisher = {Springer},
pdf = {pdf/2004/Delany2004Case.pdf},
url = {http://www.cs.tcd.ie/publications/tech-reports/reports.04/TCD-CS-2004-30.pdf}
}
@article{Doyle2004Fionn,
author = {Doyle, D{\'o}nal and Loughrey, John and Nugent, Conor and Coyle, Lorcan and Cunningham, P{\'a}draig},
title = {FIONN: A Framework for Developing CBR Systems},
journal = {Expert Update},
year = {2004},
volume = {8},
pages = {11--14},
number = {1},
pdf = {pdf/2004/Doyle2004Fionn.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1146585684938.pdf}
}
As e-commerce has become more popular, the problem of information overload has come to the fore. Recommender systems that reduce the information overload problem are becoming more common. However, the problem with many recommender systems is that they are associated with a high cost of learning customer preferences (in terms of cognitive load). We describe the Personal Travel Assistant (PTA), a flight recommender application that uses case-based reasoning (CBR) to overcome these problems. The PTA allows users to search multiple flights providers concurrently and recommends flights based on their individual travel preferences. These preferences are implicitly learned from observations of user behaviour. When the user purchases a flight, the PTA uses the selection of a preferred flight to discover and refine the user’s overall travel preferences. These preferences are stored in a user-model as sets of cases representing their interactions, which are used to provide personalised recommendations. The PTA makes recommendations taking into account the context in which the flights were offered. It uses features from the request to determine this context, e.g. the duration of the trip. We perform evaluations of contextual recommendations that support our view that user preferences change depending on the context of the session. We further improve recommendation accuracy by storing and personalising similarity measures in the user-model. The PTA alters the relative importance of features in the personal similarity measure based on implicit user feedback, e.g. increasing the importance of price at the cost of stop-over time in a multiple hop flight. We also investigate cooperative components to extend our recommendation strategies. These allow users to reuse the information learned from other users when they encounter new situations. However, these techniques are not as successful as we had hoped. We discuss these components in relation to other work on collaborative recommendation and suggest that the standard approach is unsuited to the PTA’s context-based recommendation strategy. The strength of CBR in the e-commerce domain stems from its reuse of the knowledge base associated with a particular application. Since case data may be one aspect of a company’s entire knowledge system, it is important to integrate case data easily within a company’s IT infrastructure, providing in effect a case-based view on relevant portions of the company knowledge base. We describe CBML, an XML-based Case Mark-Up Language we have developed to facilitate such integration.
@phdthesis{Coyle2004Making,
author = {Coyle, Lorcan},
title = {Making Personalised Flight Recommendations using Implicit Feedback},
school = {Trinity College Dublin},
year = {2004},
pdf = {pdf/2004/Coyle2004Making.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.05/TCD-CS-2005-32.pdf}
}
Because of the changing nature of spam, a spam filtering system that uses machine learning will need to be dynamic. This suggests that a case-based (memory-based) approach may work well. Case-Based Reasoning (CBR) is a lazy approach to machine learning where induction is delayed to run time. This means that the case base can be updated continuously and new training data is immediately available to the induction process. In this paper we present a detailed description of such a system called ECUE and evaluate design decisions concerning the case representation. We compare its performance with an alternative system that uses Naı̈ve Bayes (NB). We find that there is little to choose between the two alternatives in cross-validation tests on data sets. However, ECUE does appear to have some advantages in tracking concept drift over time.
@inproceedings{Delany2004Assessment,
author = {Delany, Sarah Jane and Cunningham, P{\'a}draig and Coyle, Lorcan},
title = {An Assessment of Case-Based Reasoning for Spam Filtering},
booktitle = {Proceedings of the Fifteenth Irish Conference on Artificial Intelligence
and Cognitive Science (AICS'2004)},
year = {2004},
editor = {McGinty, Lorraine and Crean, Brian},
pages = {9--18},
pdf = {pdf/2004/Delany2004Assessment.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.04/TCD-CS-2004-44.pdf}
}
The ranking of offers is an issue in e-commerce that has received a lot of attention in Case-Based Reasoning research. In the absence of a sales assistant, it is important to provide a facility that will bring suitable products and services to the attention of the customer. In this paper we present such a facility that is part of a Personal Travel Assistant (PTA) for booking flights online. The PTA returns a large number of offers (24 on average) and it is important to rank them to bring the most suitable to the fore. This ranking is done based on similarity to previously accepted offers. It is a characteristic of this domain that the case-base of accepted offers will be small, so the learning of appropriate feature weights is a particular challenge. We describe a process for learning personalised feature weights and present an evaluation that shows its effectiveness.
@inproceedings{Coyle2004Improving,
author = {Coyle, Lorcan and Cunningham, P{\'a}draig},
title = {Improving Recommendation Ranking by Learning Personal Feature Weights},
booktitle = {Advances in Case-Based Reasoning, 7th European Conference, ECCBR
2004 Madrid, Spain, August 30th through September 2nd, 2004, Proceedings},
year = {2004},
editor = {Calero, Pedro A. Gonz{\'a}lez and Funk, Peter},
pages = {560--572},
publisher = {Springer},
doi = {10.1007/b99702},
pdf = {pdf/2004/Coyle2004Improving.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.04/TCD-CS-2004-21.pdf}
}
As Case-Based Reasoning has matured as a discipline; the need for a standard means of representing case-based knowledge has come to the fore. While proposals exist for representing the vocabulary and the case-base knowledge containers, there are still no proposed standards for representing similarity or adaptation knowledge. In this paper we present extensions for representing similarity knowledge to CBML, an XML-based CBR language.
@inproceedings{Coyle2004Representing,
author = {Coyle, Lorcan and Doyle, D{\'o}nal and Cunningham, P{\'a}draig},
title = {Representing Similarity for CBR in XML},
booktitle = {Advances in Case-Based Reasoning, 7th European Conference, ECCBR
2004 Madrid, Spain, August 30th through September 2nd, 2004, Proceedings},
year = {2004},
editor = {Calero, Pedro A. Gonz{\'a}lez and Funk, Peter},
pages = {119--127},
publisher = {Springer},
pdf = {pdf/2004/Coyle2004Representing.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.04/TCD-CS-2004-25.pdf}
}
Intelligent software assistants are becoming more common in the e-commerce domain. We are working on a personal travel assistant. The goal of this application is to use case based reasoning to assist the user in arranging flights. It offers personalised service to its users and automatically learns their travel preferences. It stores these preferences in a user model that is directly related to the CBR process. It learns the user preferences by exploiting user feedback on sets of presented travel offers. When the user selects a preferred offer, the PTA establishes a preference ordering among the whole set. This ordering is calculated by measuring the similarity between the selected offer and each of the other offers. This ordering is used to rate these offers and store them in the user profile as cases. This ordering is also used to refine the user’s overall travel preferences by altering their personal similarity measure.
@inproceedings{Coyle2003Exploiting,
author = {Coyle, Lorcan and Cunningham, P{\'a}draig},
title = {Exploiting Re-ranking Information in a Case-Based Personal Travel
Assistant},
booktitle = {Workshop in Mixed-Initiative Case-Based Reasoning, Workshop Programme
at the Fifth International Conference on Case-Based Reasoning},
year = {2003},
editor = {Aha, David},
pages = {11-20},
pdf = {pdf/2003/Coyle2003Exploiting.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.03/TCD-CS-2003-18.pdf}
}
Case Based Reasoning has found increasing application on the Internet as a shopping assistant for e-commerce stores. The strength of CBR in this area stems from its reuse of the knowledge base associated with a particular application, thus providing an ideal way to make personalised configuration or technical information available to the Internet user. Since case data may be one aspect of a company’s entire knowledge system, it is important to integrate case data easily within a company’s IT infrastructure, providing in effect a case-based view on relevant portions of the company knowledge base. We describe CBML, an XML-based Case Mark-Up Language we have developed to facilitate such integration. We will detail the benefits of our system for industry in general in terms of extensibility, ease of reuse and interoperability. The language allows us to make the formal definition of the structure of our cases completely independent of the application code. In this way we allow the structure and definition of our cases to be described and modified easily. Such a language would also allow cases to be exchanged between heterogeneous CBR systems. As an example of how CBML might be used we describe our research on a wireless Case Based assistant for the travel market. In this application user profiles are marked up as sets of cases in CBML.
@article{Coyle2003Representing,
author = {Coyle, Lorcan and Hayes, Conor and Cunningham, P{\'a}draig},
title = {Representing Cases for CBR in XML},
journal = {Expert Update},
year = {2003},
volume = {6},
pages = {7--13},
number = {2},
pdf = {pdf/2003/Coyle2003Representing.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.02/TCD-CS-2002-63.pdf}
}
Case Based Reasoning has found increasing application on the Internet as a shopping assistant for e-commerce stores. The strength of CBR in this area stems from its reuse of the knowledge base associated with a particular application, thus providing an ideal way to make personalised configuration or technical information available to the Internet user. Since case data may be one aspect of a company’s entire knowledge system, it is important to integrate case data easily within a company’s IT infrastructure, providing in effect a CBR ’View’ on the company knowledge base. We describe CBML, an XML-based Case Mark-Up Language we have developed to facilitate such integration. We will detail the benefits of our system for industry in general in terms of extensibility, ease of reuse and interoperability. The language allows us to make the formal definition of the structure of our cases completely independent of the application code. In this way we allow the structure and definition of our cases to be described and modified easily. Such a language would also allow cases to be exchanged between heterogeneous CBR systems. As an example of how CBML might be used we describe our research on a wireless Case Based assistant for the travel market. In this application user profiles are marked up as sets of cases in CBML.
@inproceedings{Coyle2002Representing,
author = {Coyle, Lorcan and Hayes, Conor and Cunningham, P{\'a}draig},
title = {Representing Cases for CBR in XML},
booktitle = {7th UKCBR Workshop, Peterhouse, Cambridge, UK, 2002},
year = {2002},
pdf = {pdf/2002/Coyle2002Representing.pdf},
url = {http://www.csi.ucd.ie/UserFiles/publications/1193758598010.pdf}
}
This paper describes a case-based approach to user profiling in a Personal Travel assistant (based on the 1998 FIPA Travel Scenario). The approach is novel in that the user profile is made up of a set of cases capturing previous interactions rather than as a single composite case. This has the advantage that the profile is always up-to-date and also allows for the borrowing of cases from similar users when coverage is poor. Profile data is retrieved from a database in an XML format and loaded into a case-retrieval net in memory. This case-retrieval net is then used to support the two key tasks of requirements elaboration and ranking offers.
@inproceedings{Coyle2002Case,
author = {Coyle, Lorcan and Cunningham, P{\'a}draig and Hayes, Conor},
title = {A Case-Based Personal Travel Assistant for Elaborating User Requirements
and Assessing Offers},
booktitle = {Advances in Case-Based Reasoning, 6th European Conference, ECCBR
2002 Aberdeen, Scotland, UK, September 4-7, 2002, Proceedings},
year = {2002},
editor = {Craw, Susan and Preece, Alun D.},
pages = {505--518},
publisher = {Springer},
doi = {10.1007/3-540-46119-1_37},
pdf = {pdf/2002/Coyle2002Case.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.02/TCD-CS-2002-17.pdf}
}
This report details the design and implementation of an artificial world that demonstrates Darwinian evolution. This is done using an agent based modelling tool called Swarm. This world is populated with hundreds of agents that compete with one another to survive. Their behavioural attributes are coded into their genes. The initial population consists of individuals with randomly coded genes. It is hoped that by exerting Darwinian evolution on this, a population of agents, optimally suited for survival in the world will emerge.
@techreport{Coyle2001Demonstrating,
author = {Coyle, Lorcan},
title = {Demonstrating Darwinian Evolution using Swarm},
institution = {Trinity College Dublin},
year = {2001},
number = {TCD-CS-2001-47},
note = {Final Year Project.},
pdf = {pdf/2001/Coyle2001Demonstrating.pdf},
url = {https://www.cs.tcd.ie/publications/tech-reports/reports.01/TCD-CS-2001-48.pdf}
}